Typically, a solid-state laser cavity contains a solid host material that is doped with a small amount of an activator ion collectively called the laser gain medium (LGM). This LGM can be pumped by a light source of suitable frequency. The light from the pump is absorbed by the activator ion in the LGM, creating a population inversion that causes stimulated emission of coherent light. The LGM is typically placed between two or more mirrors that internally reflect the pump light and create standing waves of coherent light. Such an arrangement is called a resonator cavity. The mirrors can be in the form of external self-standing objects with the LGM between them or dielectric coatings deposited directly on faces of the LGM crystal. The output light can be in the form of continuous or pulsed emission. Coherent laser light is generally understood to have photons all having the same wavelength with the same phase and propagating along the same vector.
The LGM in a solid-state laser is typically pumped with a flashlamp or diode pump. While the pumping process can be very efficient, particularly in the case of diode pumping, the process also puts considerable strain on the crystal because localized absorption also creates localized heating due to the quantum defect inherent in any laser pumping process. Quantum defects occur when the diode pump excites the lasing ions to a state that is higher in energy than the state from which they lase (the upper lasing state). The relaxation from initial excited state to the upper lasing state typically occurs by emission of energy in the form of heat. Such emission creates local heating that can be significant, especially for high power laser applications. Since the pump beam is so focused and localized, the corresponding heating will also be localized. Such localized heating is generally not desirable since it creates a number of negative effects such as thermal lensing and lattice expansion. Such effects can cause cracking of dielectric coatings and decreased quality of the lasing beam. In many cases the waste heat has been removed by contact with a heat sink, which can be located adjacent to the LGM.
While the gain medium can be the only crystal regime of a laser cavity, solid-state lasers often employ several other single crystal regimes that serve a series of purposes including thermal management, mechanical strength increase, waveguiding, Q-switching, harmonic generation and the like. These regimes can occur in the form of a series of separate and distinct single crystals, or alternatively, adjacent layers or films that have similar lattice structures and dimensions, but with slightly different chemical compositions where the different compositions reflect the different functions. For instance, LGM have been developed that have regions that vary from one another with regard to presence and/or amount of dopant ions. These LGM generally include individual pieces that contain gradient doping or contain individual layers within each piece that can serve the desired purpose.
One variation of solid-state lasers is the thin disk design. The thin disk design is particularly suitable for high power and high beam quality. In this case a crystal consists of thicker undoped substrate such as YAG, with approximate dimensions of 1 cm2 area on a surface and 1 mm thickness. A second part of this LGM is a thin layer (perhaps 100-150 microns) of the same material doped with a suitable lasing ion such as Yb3+. The layered LGM is mounted on a reflective heat sink. The crystal is pumped typically with a diode laser from the top and the reflective heat sink reflects the pump back up through the thin layer of lasing ions. A series of parabolic mirrors can repeatedly reflect the pump back down again through the disk making maximum use of the thin layer of dopant ions to generate high power (See, e.g., U.S. Pat. No. 6,577,666 to Steffen et al., and A. Giesen, et al. (IEEE Journal of Selected Topics in Quantum Electronics 13, 2007, 598-609)). The ability of the heat sink to remove excess heat minimizes the thermal defects that can occur at higher power operation such as thermal lensing and beam distortion. In addition, the thin layer of the lasing region minimizes the degree of strain and distortion that occurs in this region. These effects lead to high power and good beam quality. Alternatively an undoped end cap can also be introduced at the top of the crystal to minimize lattice deformation of the thin, doped layer and mitigate against damage of the necessary thin film dielectric coating and to prevent lower beam quality.
Another variation in solid-state laser design employs a waveguide LGM. These typically employ a thin film of host material doped with a laser active ion. This thin film is sandwiched between two other layers typically the same host which is undoped. The layers are chosen so the thin, doped layer has a higher index of refraction than the outer cladding layers creating a total internal reflectance condition. Waveguides have many applications related to laser operations. They have been prepared by a variety of methods. For example they have been prepared by layering several pieces including cladding layers and a lasing layer that have been polished to an optical finish and heating them to induce diffusion bonding as described in U.S. Pat. No. 7,217,585 to Sumida, et al.). In another approach a waveguide device has been grown using liquid phase epitaxy in which the layers are grown from a molten flux (see, e.g., U.S. Pat. No. 5,175,787 to Gualtieri, et al.; Ferrand, et al. (Optical Materials 11 (1999) 1010-114); and Pelenc, et al. (Optical Communications 115 (1995) 491-497)). In addition there have been described a large number of beam etching methods such as reactive ion beam and ion beam implantation to generate an optically active embedded layer in waveguide crystals (see, e.g., (Pollnau, et al. (Physique 8 (2007) 123-137)).
Unfortunately, LGM like thin disks and waveguides that include multiple regimes suffer from efficiency problems due at least in part due to the interface between the crystals forming the different regimes. One common method for forming LGM with multiple regions is to use thermal bonding methods to physically attach the separate pieces together. This is also commonly referred to as diffusion bonding and a number of descriptions exist that employ this technique in addition to the waveguide descriptions mentioned above (see, e.g., Griebner, et al. (Optics Communications 1999, 164, 185-190), or U.S. Pat. Nos. 5,846,638 to Meissner, 5,441,803 to Meissner, 6,160,824 to Meissner, et al. and 7,217,585 to Sumida et al.). Similarly, the various layers of a thin disk laser are typically attached by the diffusion bonding method (see, e.g., U.S. Pat. No. 6,347,109 to Beach et al. and U.S. Pat. No. 8,165,182 to Geisen et al.). In this method, separate pieces are polished to an extremely high level of flatness and smoothness and heated to 50-90% of their melting points under pressure. The pieces are then fused together at the polished interface. The ability to fuse different types of pieces together enables the formation of gradient doping as well as waveguide devices.
An alternative method used to prepare LGM pieces with different regimes has been to grow the layers individually on appropriate substrates using the liquid phase epitaxy (LPE) method. In this case a suitable feedstock is dissolved in a high temperature melt (usually of a molten salt) and the substrate is dipped into the melt. If the solution is supersaturated then layers of the desired material can be grown on the substrate. (see, e.g., J. Hulliger, et al. (Laser Physics 1998, 8, 764-768), U.S. Pat. No. 4,810,325 to Licht, Most of such disclosure has related to semiconductors or magnetic materials but some have disclosed formation of waveguides through LPE (see, e.g., U.S. Pat. No. 4,116,530 to Bellevance, et al. U.S. Pat. No. 5,175,787 to Gualtieri U.S. Pat. No. 4,766,954 to Bierlien et al. and U.S. Pat. No. 5,150,447 to Tamada et al.)). The appropriate mixing and matching of layers grown by LPE can lead to a wide variety of potential devices that employ waveguides and gradient doping (see, e.g., Ferrand, et al. (Optical Materials 1999, 11 101-114)).
Unfortunately, such methods have failed to alleviate problems associated with a multiple regime crystal for use in lasing operations.
Hydrothermal epitaxial growth has been described previously for some oxides, particularly for incipient ferroelectrics (see, e.g., Hayashi, at al. (Mater. Sci. 2008, 43, 2342-2347) and Gleichman, et al, (Crystal Research Technology 2001, 36, 1181-1188)) and bubble memory garnets (especially Y3Fe5O12, YIG) (see, e.g., Ferrand, et al. (Mat. Res. Bull 1974, 9, 495-506) and Kolb, et al. (J. Crystal Growth 1975, 29, 29-39)). Hydrothermal processes have also been utilized to prepare single crystals that can be utilized in a variety of laser application (see, e.g., U.S. Pat. Nos. 7,563,320; 7,540,917; and 7,211,234 to Kolis, et al.).
What is needed in the art are methods for formation of LGM-based devices such as solid state lasers, including thin disk lasers, waveguides, and the like that can incorporate multiple regimes in a single crystal and thus avoid problems such as thermal distortion and coatings damage. Moreover, a method based upon hydrothermal epitaxial growth could present a variety of benefits to the art.